Home/Notes/Normalization
Board Exam Notes

Normalization Notes

Questions

3–5 questions in university semester exams

Difficulty

Medium-Hard

Importance

Core topic for university DBMS papers

Overview

Normalization is the systematic process of organizing data in a database to minimize redundancy and eliminate undesirable anomalies like insertion, update, and deletion issues. It is a fundamental concept in DBMS exams that tests your ability to decompose complex relations into smaller, well-structured tables using functional dependencies.

Functional Dependencies (FD)

Functional dependencies define the relationship between attributes in a relation, specifically stating that a value of one attribute uniquely determines the value of another. Understanding the closure of an attribute set is crucial for determining candidate keys and identifying the level of normalization.

  • A -> B means A functionally determines B
  • Armstrong's Axioms: Reflexivity, Augmentation, Transitivity
  • Trivial FD: Y is a subset of X
  • Closure of X (X+) represents all attributes functionally dependent on X
  • Candidate Key: Minimal set of attributes whose closure is the set of all attributes in the relation

Normal Forms: 1NF, 2NF, and 3NF

Normalization moves from 1NF to 3NF by systematically removing partial and transitive dependencies. 1NF requires atomic values, 2NF removes partial functional dependencies, and 3NF ensures no non-prime attribute is transitively dependent on the primary key.

  • 1NF: Atomic values, no repeating groups
  • 2NF: Relation in 1NF and no partial dependency (non-prime attributes fully dependent on candidate key)
  • 3NF: Relation in 2NF and no transitive dependency
  • Anomalies: Insertion, Update, and Deletion anomalies in unnormalized tables
  • Prime attribute: Any attribute that is part of some candidate key

BCNF and Lossless Decomposition

BCNF is a stricter version of 3NF, addressing anomalies that occur when multiple overlapping candidate keys exist. Lossless decomposition ensures that when a relation is split, the original data can be reconstructed via natural joins without generating spurious tuples.

  • BCNF Condition: For every FD X->Y, X must be a superkey
  • Every BCNF relation is in 3NF, but not every 3NF is in BCNF
  • Lossless Join property: R1 INTERSECT R2 -> R1 or R1 INTERSECT R2 -> R2
  • Dependency Preservation: Property where all original FDs are covered by the decomposed schemas
  • It is always possible to achieve 3NF with lossless join and dependency preservation

Formula Sheet

X+ (Attribute Closure calculation)

R = R1 U R2 (Decomposition rule)

R1 Intersection R2 -> R1 (Lossless Join condition)

Exam Tip

Always find all candidate keys first using attribute closure before attempting to identify which normal form a relation satisfies.

Common Mistakes

  • Confusing partial dependency (part of a key) with transitive dependency (key -> non-prime -> non-prime)
  • Forgetting to check for all candidate keys when testing for BCNF
  • Assuming that because a decomposition is lossless, it is also dependency-preserving

More Revision Notes

Ready to test yourself?

Play topic-wise Normalization questions in Aspirant Arcade — gamified MCQ practice.

Download Free